Quantitative Economics 2023年第1期

Volume 14,?Issue 1, January 2023

?

?

——更多動態(tài)，請持續(xù)關(guān)注gzh：理想主義的百年孤獨(dú)

?

1.Full-information estimation of heterogeneous agent models using macro and micro data

利用宏觀和微觀數(shù)據(jù)的異構(gòu)代理模型的全信息估計(jì)

Laura Liu,??Mikkel Plagborg-M?ller

We develop a generally applicable full-information inference method for heterogeneous agent models, combining aggregate time series data and repeated cross-sections of micro data. To handle unobserved aggregate state variables that affect cross-sectional distributions, we compute a numerically unbiased estimate of the model-implied likelihood function. Employing the likelihood estimate in a Markov Chain Monte Carlo algorithm, we obtain fully efficient and valid Bayesian inference. Evaluation of the micro part of the likelihood lends itself naturally to parallel computing. Numerical illustrations in models with heterogeneous households or firms demonstrate that the proposed full-information method substantially sharpens inference relative to using only macro data, and for some parameters micro data is essential for identification.

我們開發(fā)了一種適用于異構(gòu)代理模型的全信息推理方法，結(jié)合了聚合的時(shí)間序列數(shù)據(jù)和微觀數(shù)據(jù)的重復(fù)截面。為了處理影響截面分布的不可觀察的聚集狀態(tài)變量，我們計(jì)算了模型隱含似然函數(shù)的數(shù)值無偏估計(jì)。利用馬爾可夫鏈蒙特卡羅算法中的似然估計(jì)，我們得到了完全有效的貝葉斯推斷。對可能性的微觀部分的評估自然適合于并行計(jì)算。在包含不同家庭或企業(yè)的模型中，數(shù)值插圖表明，與僅使用宏觀數(shù)據(jù)相比，所提出的全信息方法大大增強(qiáng)了推斷能力，對于某些參數(shù)，微觀數(shù)據(jù)是識別的必要條件。

?

2.Permutation-based tests for discontinuities in event studies

事件研究中基于排列的不連續(xù)性檢驗(yàn)

Federico A. Bugni,??Jia Li,??Qiyuan Li

We propose using a permutation test to detect discontinuities in an underlying economic model at a known cutoff point. Relative to the existing literature, we show that this test is well suited for event studies based on time-series data. The test statistic measures the distance between the empirical distribution functions of observed data in two local subsamples on the two sides of the cutoff. Critical values are computed via a standard permutation algorithm. Under a high-level condition that the observed data can be coupled by a collection of conditionally independent variables, we establish the asymptotic validity of the permutation test, allowing the sizes of the local subsamples to be either be fixed or grow to infinity. In the latter case, we also establish that the permutation test is consistent. We demonstrate that our high-level condition can be verified in a broad range of problems in the infill asymptotic time-series setting, which justifies using the permutation test to detect jumps in economic variables such as volatility, trading activity, and liquidity. These potential applications are illustrated in an empirical case study for selected FOMC announcements during the ongoing COVID-19 pandemic.

我們建議使用排列檢驗(yàn)來檢測潛在經(jīng)濟(jì)模型在一個(gè)已知的截止點(diǎn)上的不連續(xù)性。相對于現(xiàn)有的文獻(xiàn)，我們表明這種測試非常適合基于時(shí)間序列數(shù)據(jù)的事件研究。檢驗(yàn)統(tǒng)計(jì)量測量了兩個(gè)局部子樣本中觀測數(shù)據(jù)的經(jīng)驗(yàn)分布函數(shù)在斷裂帶兩側(cè)的距離。臨界值是通過標(biāo)準(zhǔn)排列算法計(jì)算的。在觀測數(shù)據(jù)可以被一組條件自變量耦合的高階條件下，我們建立了置換檢驗(yàn)的漸近有效性，允許局部子樣本的大小固定或增長到無窮。在后一種情況下，我們還確定置換檢驗(yàn)是一致的。我們證明，我們的高水平條件可以在填充漸近時(shí)間序列設(shè)置中的廣泛問題中得到驗(yàn)證，這證明了使用排列檢驗(yàn)來檢測經(jīng)濟(jì)變量的跳躍，如波動性、交易活動和流動性。在COVID-19大流行期間，對選定的FOMC公告進(jìn)行的實(shí)證案例研究說明了這些潛在的應(yīng)用。

?

3.Random utility and limited consideration

隨機(jī)效用和有限的斟酌

Victor H. Aguiar,??Maria Jose Boccardi,??Nail Kashaev,??Jeongbin Kim

The random utility model (RUM, McFadden and Richter (1990)) has been the standard tool to describe the behavior of a population of decision makers. RUM assumes that decision makers behave as if they maximize a rational preference over a choice set. This assumption may fail when consideration of all alternatives is costly. We provide a theoretical and statistical framework that unifies well-known models of random (limited) consideration and generalizes them to allow for preference heterogeneity. We apply this methodology in a novel stochastic choice data set that we collected in a large-scale online experiment. Our data set is unique since it exhibits both choice set and (attention) frame variation. We run a statistical survival race between competing models of random consideration and RUM. We find that RUM cannot explain the population behavior. In contrast, we cannot reject the hypothesis that decision makers behave according to the logit attention model (Brady and Rehbeck (2016)).

隨機(jī)效用模型(RUM, McFadden and Richter, 1990)一直是描述決策者群體行為的標(biāo)準(zhǔn)工具。朗姆假設(shè)決策者的行為就像他們在選擇集上最大化理性偏好一樣。當(dāng)考慮所有替代方案代價(jià)高昂時(shí)，這一假設(shè)可能會失敗。我們提供了一個(gè)理論和統(tǒng)計(jì)框架，統(tǒng)一了眾所周知的隨機(jī)(有限)考慮模型，并對它們進(jìn)行了一般化處理，以考慮偏好異質(zhì)性。我們將這種方法應(yīng)用于一個(gè)新的隨機(jī)選擇數(shù)據(jù)集，我們在一個(gè)大規(guī)模的在線實(shí)驗(yàn)中收集。我們的數(shù)據(jù)集是獨(dú)特的，因?yàn)樗瑫r(shí)展示了選擇集和(注意力)幀變化。我們在隨機(jī)考慮模型和朗姆酒模型之間進(jìn)行了一場統(tǒng)計(jì)上的生存競賽。我們發(fā)現(xiàn)朗姆酒不能解釋種群行為。相反，我們不能拒絕“決策者行為遵循logit注意力模型”的假設(shè)(Brady and Rehbeck(2016))。

?

4.Forecasting with a panel Tobit model

使用面板Tobit模型進(jìn)行預(yù)測

Laura Liu,??Hyungsik Roger Moon,??Frank Schorfheide

We use a dynamic panel Tobit model with heteroskedasticity to generate forecasts for a large cross-section of short time series of censored observations. Our fully Bayesian approach allows us to flexibly estimate the cross-sectional distribution of heterogeneous coefficients and then implicitly use this distribution as prior to construct Bayes forecasts for the individual time series. In addition to density forecasts, we construct set forecasts that explicitly target the average coverage probability for the cross-section. We present a novel application in which we forecast bank-level loan charge-off rates for small banks.

我們使用具有異方差的動態(tài)面板Tobit模型對截尾觀測的短時(shí)間序列的大截面產(chǎn)生預(yù)測。我們的完全貝葉斯方法允許我們靈活地估計(jì)異質(zhì)系數(shù)的截面分布，然后隱式地使用該分布作為之前構(gòu)建單個(gè)時(shí)間序列的貝葉斯預(yù)測。除了密度預(yù)測，我們構(gòu)建了明確針對橫截面平均覆蓋概率的集合預(yù)測。我們提出了一個(gè)新的應(yīng)用，在該應(yīng)用中，我們預(yù)測了小型銀行的銀行級貸款沖銷利率。

?

5.Monetary policy, external instruments, and heteroskedasticity

貨幣政策、外部工具變量與異方差

Thore Schlaak,??Malte Rieth,??Maximilian Podstawski

We develop a structural vector autoregressive framework that combines external instruments and heteroskedasticity for identification of monetary policy shocks. We show that exploiting both types of information sharpens structural inference, allows testing the relevance and exogeneity condition for instruments separately using likelihood ratio tests, and facilitates the economic interpretation of the structural shock of interest. We test alternative instruments and find that narrative and model-based measures are valid, while high-frequency data instruments show signs of invalidity. Finally, we document that monetary shocks identified with both a valid instrument and heteroskedasticity have larger effects on production and prices than monetary shocks identified via an instrument only.

我們開發(fā)了一個(gè)結(jié)合外部工具和異方差的結(jié)構(gòu)向量自回歸框架來識別貨幣政策沖擊。研究表明，利用這兩種信息可以強(qiáng)化結(jié)構(gòu)推斷，允許使用似然比檢驗(yàn)分別檢驗(yàn)工具的相關(guān)性和外生性條件，并有助于對利益結(jié)構(gòu)沖擊的經(jīng)濟(jì)解釋。我們測試了其他工具，發(fā)現(xiàn)敘述性和基于模型的措施是有效的，而高頻數(shù)據(jù)工具顯示出無效的跡象。最后，我們發(fā)現(xiàn)，與僅通過工具識別的貨幣沖擊相比，有效工具和異方差識別的貨幣沖擊對生產(chǎn)和價(jià)格的影響更大。

?

6.Pareto extrapolation: An analytical framework for studying tail inequality

帕累托外推:研究尾不等式的分析框架

émilien Gouin-Bonenfant,??Alexis Akira Toda

We develop an analytical framework designed to solve and analyze heterogeneous-agent models that endogenously generate fat-tailed wealth distributions. We exploit the asymptotic linearity of policy functions and the analytical characterization of the Pareto exponent to augment the conventional solution algorithm with a theory of the tail. Our framework allows for a precise understanding of the very top of the wealth distribution (e.g., analytical expressions for top wealth shares, type distribution in the tail, and transition probabilities in and out of the tail) in addition to delivering improved accuracy and speed.

我們開發(fā)了一個(gè)分析框架，旨在解決和分析內(nèi)生地產(chǎn)生胖尾財(cái)富分布的異質(zhì)性代理模型。我們利用策略函數(shù)的漸近線性和帕累托指數(shù)的解析表征，以尾巴理論來增強(qiáng)傳統(tǒng)的解決算法。我們的框架允許精確理解財(cái)富分布的頂部(例如，頂級財(cái)富份額的解析表達(dá)式、尾部的類型分布以及尾部內(nèi)外的轉(zhuǎn)移概率)，此外還提高了準(zhǔn)確性和速度。

?

7.Testing unified growth theory: Technological progress and the child quantity-quality tradeoff

檢驗(yàn)統(tǒng)一成長理論:技術(shù)進(jìn)步與兒童數(shù)量與質(zhì)量的權(quán)衡

Jakob Madsen,??Holger Strulik

?

A core mechanism of unified growth theory is that accelerating technological progress induces mass education and, through interaction with child quantity-quality substitution, a decline in fertility. Using unique new data for 21 OECD countries over the period 1750–2000, we test, for the first time, the validity of this core mechanism of unified growth theory. We measure a country's technological progress as patents per capita, R&D intensity, and investment in machinery, equipment, and intellectual property products. While controlling for confounders, such as income growth, mortality, and the gender wage gap, we establish (1) a significant impact of technological progress on education (positive) and fertility (negative); (2) that accelerating technological progress stimulated the fertility transition; and (3) that the baseline results are supported in 2SLS regressions using genetic-distance weighted foreign patent-intensity, compulsory schooling years, and minimum working age as instruments.

統(tǒng)一增長理論的一個(gè)核心機(jī)制是，加速技術(shù)進(jìn)步導(dǎo)致了大眾教育，并通過與兒童數(shù)量-質(zhì)量替代的互動，導(dǎo)致了生育率的下降。利用21個(gè)經(jīng)合組織國家1750—2000年的獨(dú)特新數(shù)據(jù)，我們首次檢驗(yàn)了統(tǒng)一增長理論這一核心機(jī)制的有效性。我們用人均專利、研發(fā)強(qiáng)度、機(jī)械設(shè)備和知識產(chǎn)權(quán)產(chǎn)品投資等指標(biāo)來衡量一個(gè)國家的技術(shù)進(jìn)步。在控制了收入增長、死亡率和性別工資差距等混雜因素的同時(shí)，我們確定了(1)技術(shù)進(jìn)步對教育(正)和生育率(負(fù))的顯著影響;(2)加速的技術(shù)進(jìn)步促進(jìn)了生育轉(zhuǎn)型;(3)以基因距離加權(quán)的外國專利密集度、義務(wù)教育年限和最低工作年齡為工具進(jìn)行2SLS回歸，基線結(jié)果得到支持。

?

8.Borrowing into debt crises

舉債應(yīng)對債務(wù)危機(jī)

Radoslaw Paluszynski,??Georgios Stefanidis

Quantitative models of sovereign default predict that governments reduce borrowing during recessions to avoid debt crises. A prominent implication of this behavior is that the resulting interest rate spread volatility is counterfactually low. We propose that governments borrow into debt crises because of frictions in the adjustment of their expenditures. We develop a model of government good production, which uses public employment and intermediate consumption as inputs. The inputs have varying degrees of downward rigidity, which means that it is costly to reduce them. Facing an adverse income shock, the government borrows to smooth out the reduction in public employment, which results in increasing debt and higher spread. We quantify this rigidity using the OECD Government Accounts data and show that it explains about 70% of the missing bond spread volatility.

主權(quán)違約的定量模型預(yù)測，政府會在衰退期間減少借款，以避免債務(wù)危機(jī)。這種行為的一個(gè)突出含義是，由此產(chǎn)生的利差波動是反事實(shí)的低。我們建議，政府在債務(wù)危機(jī)中舉債是因?yàn)樵谡{(diào)整支出方面存在摩擦。我們建立了一個(gè)以公共就業(yè)和中間消費(fèi)為投入的政府產(chǎn)品生產(chǎn)模型。投入具有不同程度的向下剛性，這意味著降低投入的成本較高。面對不利的收入沖擊，政府借款以平滑公共就業(yè)的減少，這導(dǎo)致債務(wù)增加和利差上升。我們使用經(jīng)合組織(OECD)的政府賬戶數(shù)據(jù)對這種剛性進(jìn)行了量化，并表明它解釋了約70%的債券利差波動缺失。

?

?

9.The demographic consequences of sex-selection technology

性別選擇技術(shù)對人口統(tǒng)計(jì)學(xué)的影響

Qi Li,??Juan Pantano

Over the last several years, highly accurate methods of sex selection before conception have been developed. Given that strong preferences for sex variety in offspring have been documented for the U.S., we ask what the demographic consequences of sex-selection technology could be. Lacking variation across space and time in access to this technology, we estimate a dynamic programming model of fertility decisions with microdata on fertility histories. We leverage the quasi-experimental variation inherent in the random determination of sex to identify the key structural parameter characterizing preferences for sex variety in offspring. We then simulate the introduction of this technology. While this technology can reduce fertility by allowing parents to efficiently reach their preferred sex mix, it could also increase it. This is because without this technology, many parents may opt not to have another baby given the uncertainty about its sex. Results suggest that these two effects operate simultaneously, but on net, sex-selection technology ends up reducing the average family size among married women by less than 2% in the steady state, a much smaller decline than the one that would be predicted by alternative methods.

在過去的幾年中，已經(jīng)發(fā)展出了在受孕前進(jìn)行性別選擇的高度準(zhǔn)確的方法。鑒于美國對后代性別多樣性的強(qiáng)烈偏好已經(jīng)被記錄在案，我們想知道性別選擇技術(shù)可能會給人口統(tǒng)計(jì)學(xué)帶來什么后果。由于缺乏對這一技術(shù)的不同時(shí)空的獲取，我們利用生育歷史的微觀數(shù)據(jù)來估計(jì)生育決策的動態(tài)規(guī)劃模型。我們利用性別隨機(jī)決定中固有的準(zhǔn)實(shí)驗(yàn)變異來識別表征后代性別多樣性偏好的關(guān)鍵結(jié)構(gòu)參數(shù)。然后，我們模擬該技術(shù)的引入。雖然這項(xiàng)技術(shù)可以通過讓父母有效地達(dá)到他們喜歡的性別組合來降低生育率，但它也可以提高生育率。這是因?yàn)槿绻麤]有這項(xiàng)技術(shù)，許多父母可能會選擇不再生一個(gè)孩子，因?yàn)樗麄儾淮_定孩子的性別。結(jié)果表明，這兩種影響同時(shí)起作用，但從整體上看，性別選擇技術(shù)最終使已婚婦女在穩(wěn)定狀態(tài)下的平均家庭規(guī)模減少了不到2%，這比其他方法預(yù)測的降幅要小得多。

?

10.Gender, competition, and performance: Evidence from chess players

性別、競爭和表現(xiàn):來自國際象棋選手的證據(jù)

Peter Backus,??Maria Cubel,??Matej Guid,??Santiago Sánchez-Pagés,??Enrique López Ma?as

This paper studies gender differences in performance in a male-dominated competitive environment chess tournaments. We find that the gender composition of chess games affects the behaviors of both men and women in ways that worsen the outcomes for women. Using a unique measure of within-game quality of play, we show that women make more mistakes when playing against men. Men, however, play equally well against male and female opponents. We also find that men persist longer before losing to women. Our results shed some light on the behavioral changes that lead to differential outcomes when the gender composition of competitions varies.

本文研究在男性主導(dǎo)的國際象棋比賽的競爭環(huán)境中表現(xiàn)的性別差異。我們發(fā)現(xiàn)，國際象棋游戲的性別構(gòu)成會影響男性和女性的行為，從而惡化女性的結(jié)果。使用一種獨(dú)特的游戲質(zhì)量測量方法，我們發(fā)現(xiàn)女性在與男性游戲時(shí)犯的錯(cuò)誤更多。然而，男性在與男性和女性對手的比賽中表現(xiàn)得一樣好。我們還發(fā)現(xiàn)，男性在輸給女性之前堅(jiān)持的時(shí)間更長。我們的研究結(jié)果在一定程度上揭示了當(dāng)比賽的性別構(gòu)成不同時(shí)，導(dǎo)致不同結(jié)果的行為變化。